Training in progress, epoch 20, checkpoint
Browse files- checkpoint-10580/adapter_config.json +1 -1
- checkpoint-10580/adapter_model.bin +3 -0
- checkpoint-10580/added_tokens.json +3 -3
- checkpoint-10580/optimizer.pt +2 -2
- checkpoint-10580/rng_state_0.pth +1 -1
- checkpoint-10580/rng_state_1.pth +1 -1
- checkpoint-10580/rng_state_2.pth +1 -1
- checkpoint-10580/rng_state_3.pth +1 -1
- checkpoint-10580/special_tokens_map.json +4 -12
- checkpoint-10580/tokenizer.json +0 -0
- checkpoint-10580/tokenizer.model +3 -0
- checkpoint-10580/tokenizer_config.json +17 -20
- checkpoint-10580/trainer_state.json +317 -436
- checkpoint-10580/training_args.bin +2 -2
checkpoint-10580/adapter_config.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"auto_mapping": null,
|
| 3 |
-
"base_model_name_or_path": "
|
| 4 |
"bias": "none",
|
| 5 |
"fan_in_fan_out": false,
|
| 6 |
"inference_mode": true,
|
|
|
|
| 1 |
{
|
| 2 |
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
|
| 4 |
"bias": "none",
|
| 5 |
"fan_in_fan_out": false,
|
| 6 |
"inference_mode": true,
|
checkpoint-10580/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6e0a8a2319286bf4f2f3d5601f23391e4245f29d35aa8976cbca3ec48bd32667
|
| 3 |
+
size 143269386
|
checkpoint-10580/added_tokens.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"</s>": 2,
|
| 3 |
+
"<s>": 1,
|
| 4 |
+
"<unk>": 0
|
| 5 |
}
|
checkpoint-10580/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e6845fc930c455be6e2a4f3d608ef75bd044ef5197c2a9f94d51a6fe89ad2cf0
|
| 3 |
+
size 286585234
|
checkpoint-10580/rng_state_0.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15024
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3ac3ea29cda3836f11acb9479e1eae08ba089b6b4a646cf00c77177ddd3183e3
|
| 3 |
size 15024
|
checkpoint-10580/rng_state_1.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15024
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e6e0298fa228a35518a1cbb6fdf26a76dff87999edf0a6698f2980e18c730eae
|
| 3 |
size 15024
|
checkpoint-10580/rng_state_2.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15024
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:93ece22d85e620b66860f8a7ed66106457a6b3419fa23c5e63c6237d01f26776
|
| 3 |
size 15024
|
checkpoint-10580/rng_state_3.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15024
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:66214b24fdb59ad45d26bd599486934dd941439daa4c2b96b0a97e55ad82e5f3
|
| 3 |
size 15024
|
checkpoint-10580/special_tokens_map.json
CHANGED
|
@@ -1,14 +1,6 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
"eos_token": {
|
| 7 |
-
"content": "<|endoftext|>",
|
| 8 |
-
"lstrip": false,
|
| 9 |
-
"normalized": false,
|
| 10 |
-
"rstrip": false,
|
| 11 |
-
"single_word": false
|
| 12 |
-
},
|
| 13 |
-
"pad_token": "<|endoftext|>"
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"eos_token": "</s>",
|
| 4 |
+
"pad_token": "</s>",
|
| 5 |
+
"unk_token": "<unk>"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
}
|
checkpoint-10580/tokenizer.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-10580/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
|
| 3 |
+
size 499723
|
checkpoint-10580/tokenizer_config.json
CHANGED
|
@@ -1,24 +1,23 @@
|
|
| 1 |
{
|
| 2 |
-
"add_prefix_space": false,
|
| 3 |
"added_tokens_decoder": {
|
| 4 |
-
"
|
| 5 |
-
"content": "
|
| 6 |
"lstrip": false,
|
| 7 |
"normalized": false,
|
| 8 |
"rstrip": false,
|
| 9 |
"single_word": false,
|
| 10 |
"special": true
|
| 11 |
},
|
| 12 |
-
"
|
| 13 |
-
"content": "
|
| 14 |
"lstrip": false,
|
| 15 |
"normalized": false,
|
| 16 |
"rstrip": false,
|
| 17 |
"single_word": false,
|
| 18 |
"special": true
|
| 19 |
},
|
| 20 |
-
"
|
| 21 |
-
"content": "
|
| 22 |
"lstrip": false,
|
| 23 |
"normalized": false,
|
| 24 |
"rstrip": false,
|
|
@@ -26,18 +25,16 @@
|
|
| 26 |
"special": true
|
| 27 |
}
|
| 28 |
},
|
| 29 |
-
"additional_special_tokens": [
|
| 30 |
-
|
| 31 |
-
"<|im_end|>"
|
| 32 |
-
],
|
| 33 |
-
"bos_token": null,
|
| 34 |
-
"chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 35 |
"clean_up_tokenization_spaces": false,
|
| 36 |
-
"eos_token": "
|
| 37 |
-
"
|
| 38 |
-
"model_max_length":
|
| 39 |
-
"pad_token": "
|
| 40 |
-
"
|
| 41 |
-
"
|
| 42 |
-
"
|
|
|
|
|
|
|
| 43 |
}
|
|
|
|
| 1 |
{
|
|
|
|
| 2 |
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<unk>",
|
| 5 |
"lstrip": false,
|
| 6 |
"normalized": false,
|
| 7 |
"rstrip": false,
|
| 8 |
"single_word": false,
|
| 9 |
"special": true
|
| 10 |
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<s>",
|
| 13 |
"lstrip": false,
|
| 14 |
"normalized": false,
|
| 15 |
"rstrip": false,
|
| 16 |
"single_word": false,
|
| 17 |
"special": true
|
| 18 |
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "</s>",
|
| 21 |
"lstrip": false,
|
| 22 |
"normalized": false,
|
| 23 |
"rstrip": false,
|
|
|
|
| 25 |
"special": true
|
| 26 |
}
|
| 27 |
},
|
| 28 |
+
"additional_special_tokens": [],
|
| 29 |
+
"bos_token": "<s>",
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
"clean_up_tokenization_spaces": false,
|
| 31 |
+
"eos_token": "</s>",
|
| 32 |
+
"legacy": false,
|
| 33 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 34 |
+
"pad_token": "</s>",
|
| 35 |
+
"padding_side": "left",
|
| 36 |
+
"sp_model_kwargs": {},
|
| 37 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 38 |
+
"unk_token": "<unk>",
|
| 39 |
+
"use_default_system_prompt": true
|
| 40 |
}
|
checkpoint-10580/trainer_state.json
CHANGED
|
@@ -9,97 +9,87 @@
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm": 0.3393945097923279,
|
| 14 |
"learning_rate": 0.0003,
|
| 15 |
-
"loss": 1.
|
| 16 |
"step": 100
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm": 8017.20361328125,
|
| 21 |
"learning_rate": 0.0003,
|
| 22 |
-
"loss":
|
| 23 |
"step": 200
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm": 9146.7236328125,
|
| 28 |
"learning_rate": 0.0003,
|
| 29 |
-
"loss":
|
| 30 |
"step": 300
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm": 574.934326171875,
|
| 35 |
"learning_rate": 0.0003,
|
| 36 |
-
"loss":
|
| 37 |
"step": 400
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm": 25128.55859375,
|
| 42 |
"learning_rate": 0.0003,
|
| 43 |
-
"loss":
|
| 44 |
"step": 500
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 1.0,
|
| 48 |
-
"eval_accuracy": 0.
|
| 49 |
-
"eval_loss":
|
| 50 |
-
"eval_runtime":
|
| 51 |
-
"eval_samples_per_second":
|
| 52 |
-
"eval_steps_per_second":
|
| 53 |
"step": 529
|
| 54 |
},
|
| 55 |
{
|
| 56 |
"epoch": 1.0,
|
| 57 |
-
"eval_exact_match": 0.
|
| 58 |
-
"eval_f1":
|
| 59 |
"step": 529
|
| 60 |
},
|
| 61 |
{
|
| 62 |
-
"epoch": 1.
|
| 63 |
-
"grad_norm": 1332.8885498046875,
|
| 64 |
"learning_rate": 0.0003,
|
| 65 |
-
"loss":
|
| 66 |
"step": 600
|
| 67 |
},
|
| 68 |
{
|
| 69 |
-
"epoch": 1.
|
| 70 |
-
"grad_norm": 3666.881591796875,
|
| 71 |
"learning_rate": 0.0003,
|
| 72 |
-
"loss":
|
| 73 |
"step": 700
|
| 74 |
},
|
| 75 |
{
|
| 76 |
-
"epoch": 1.
|
| 77 |
-
"grad_norm": 356847.0,
|
| 78 |
"learning_rate": 0.0003,
|
| 79 |
-
"loss":
|
| 80 |
"step": 800
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"epoch": 1.
|
| 84 |
-
"grad_norm": 1061.877685546875,
|
| 85 |
"learning_rate": 0.0003,
|
| 86 |
-
"loss":
|
| 87 |
"step": 900
|
| 88 |
},
|
| 89 |
{
|
| 90 |
-
"epoch": 1.
|
| 91 |
-
"grad_norm": 46670.3046875,
|
| 92 |
"learning_rate": 0.0003,
|
| 93 |
-
"loss":
|
| 94 |
"step": 1000
|
| 95 |
},
|
| 96 |
{
|
| 97 |
"epoch": 2.0,
|
| 98 |
-
"eval_accuracy": 0.
|
| 99 |
-
"eval_loss":
|
| 100 |
-
"eval_runtime":
|
| 101 |
-
"eval_samples_per_second":
|
| 102 |
-
"eval_steps_per_second":
|
| 103 |
"step": 1058
|
| 104 |
},
|
| 105 |
{
|
|
@@ -109,204 +99,183 @@
|
|
| 109 |
"step": 1058
|
| 110 |
},
|
| 111 |
{
|
| 112 |
-
"epoch": 2.
|
| 113 |
-
"grad_norm": 2.0727381706237793,
|
| 114 |
"learning_rate": 0.0003,
|
| 115 |
-
"loss":
|
| 116 |
"step": 1100
|
| 117 |
},
|
| 118 |
{
|
| 119 |
-
"epoch": 2.
|
| 120 |
-
"grad_norm": 1.8185893297195435,
|
| 121 |
"learning_rate": 0.0003,
|
| 122 |
-
"loss":
|
| 123 |
"step": 1200
|
| 124 |
},
|
| 125 |
{
|
| 126 |
-
"epoch": 2.
|
| 127 |
-
"grad_norm": 1620.93701171875,
|
| 128 |
"learning_rate": 0.0003,
|
| 129 |
-
"loss":
|
| 130 |
"step": 1300
|
| 131 |
},
|
| 132 |
{
|
| 133 |
-
"epoch": 2.
|
| 134 |
-
"grad_norm": 1.0691380500793457,
|
| 135 |
"learning_rate": 0.0003,
|
| 136 |
-
"loss":
|
| 137 |
"step": 1400
|
| 138 |
},
|
| 139 |
{
|
| 140 |
-
"epoch": 2.
|
| 141 |
-
"grad_norm": 1.0605100393295288,
|
| 142 |
"learning_rate": 0.0003,
|
| 143 |
-
"loss":
|
| 144 |
"step": 1500
|
| 145 |
},
|
| 146 |
{
|
| 147 |
"epoch": 3.0,
|
| 148 |
-
"eval_accuracy": 0.
|
| 149 |
-
"eval_loss":
|
| 150 |
-
"eval_runtime":
|
| 151 |
-
"eval_samples_per_second":
|
| 152 |
-
"eval_steps_per_second":
|
| 153 |
"step": 1587
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 3.0,
|
| 157 |
"eval_exact_match": 0.0,
|
| 158 |
-
"eval_f1": 0.
|
| 159 |
"step": 1587
|
| 160 |
},
|
| 161 |
{
|
| 162 |
-
"epoch": 3.
|
| 163 |
-
"grad_norm": 1967676800.0,
|
| 164 |
"learning_rate": 0.0003,
|
| 165 |
-
"loss":
|
| 166 |
"step": 1600
|
| 167 |
},
|
| 168 |
{
|
| 169 |
-
"epoch": 3.
|
| 170 |
-
"grad_norm": 83918778793984.0,
|
| 171 |
"learning_rate": 0.0003,
|
| 172 |
-
"loss":
|
| 173 |
"step": 1700
|
| 174 |
},
|
| 175 |
{
|
| 176 |
-
"epoch": 3.
|
| 177 |
-
"grad_norm": 12349449.0,
|
| 178 |
"learning_rate": 0.0003,
|
| 179 |
-
"loss":
|
| 180 |
"step": 1800
|
| 181 |
},
|
| 182 |
{
|
| 183 |
-
"epoch": 3.
|
| 184 |
-
"grad_norm": 52939337728.0,
|
| 185 |
"learning_rate": 0.0003,
|
| 186 |
-
"loss":
|
| 187 |
"step": 1900
|
| 188 |
},
|
| 189 |
{
|
| 190 |
-
"epoch": 3.
|
| 191 |
-
"grad_norm": 1381182.25,
|
| 192 |
"learning_rate": 0.0003,
|
| 193 |
-
"loss":
|
| 194 |
"step": 2000
|
| 195 |
},
|
| 196 |
{
|
| 197 |
-
"epoch": 3.
|
| 198 |
-
"grad_norm": 105.80550384521484,
|
| 199 |
"learning_rate": 0.0003,
|
| 200 |
-
"loss":
|
| 201 |
"step": 2100
|
| 202 |
},
|
| 203 |
{
|
| 204 |
"epoch": 4.0,
|
| 205 |
-
"eval_accuracy": 0.
|
| 206 |
-
"eval_loss":
|
| 207 |
-
"eval_runtime":
|
| 208 |
-
"eval_samples_per_second":
|
| 209 |
-
"eval_steps_per_second":
|
| 210 |
"step": 2116
|
| 211 |
},
|
| 212 |
{
|
| 213 |
"epoch": 4.0,
|
| 214 |
-
"eval_exact_match":
|
| 215 |
-
"eval_f1":
|
| 216 |
"step": 2116
|
| 217 |
},
|
| 218 |
{
|
| 219 |
-
"epoch": 4.
|
| 220 |
-
"grad_norm": 16.110862731933594,
|
| 221 |
"learning_rate": 0.0003,
|
| 222 |
-
"loss":
|
| 223 |
"step": 2200
|
| 224 |
},
|
| 225 |
{
|
| 226 |
-
"epoch": 4.
|
| 227 |
-
"grad_norm": 4.362427234649658,
|
| 228 |
"learning_rate": 0.0003,
|
| 229 |
-
"loss":
|
| 230 |
"step": 2300
|
| 231 |
},
|
| 232 |
{
|
| 233 |
-
"epoch": 4.
|
| 234 |
-
"grad_norm": 18039.095703125,
|
| 235 |
"learning_rate": 0.0003,
|
| 236 |
-
"loss":
|
| 237 |
"step": 2400
|
| 238 |
},
|
| 239 |
{
|
| 240 |
-
"epoch": 4.
|
| 241 |
-
"grad_norm": 2.3812594051160474e+17,
|
| 242 |
"learning_rate": 0.0003,
|
| 243 |
-
"loss":
|
| 244 |
"step": 2500
|
| 245 |
},
|
| 246 |
{
|
| 247 |
-
"epoch": 4.
|
| 248 |
-
"grad_norm": 3.9549436569213867,
|
| 249 |
"learning_rate": 0.0003,
|
| 250 |
-
"loss":
|
| 251 |
"step": 2600
|
| 252 |
},
|
| 253 |
{
|
| 254 |
"epoch": 5.0,
|
| 255 |
-
"eval_accuracy": 0.
|
| 256 |
-
"eval_loss":
|
| 257 |
-
"eval_runtime":
|
| 258 |
-
"eval_samples_per_second":
|
| 259 |
-
"eval_steps_per_second":
|
| 260 |
"step": 2645
|
| 261 |
},
|
| 262 |
{
|
| 263 |
"epoch": 5.0,
|
| 264 |
"eval_exact_match": 0.0,
|
| 265 |
-
"eval_f1": 0.
|
| 266 |
"step": 2645
|
| 267 |
},
|
| 268 |
{
|
| 269 |
-
"epoch": 5.
|
| 270 |
-
"grad_norm": 120114.2421875,
|
| 271 |
"learning_rate": 0.0003,
|
| 272 |
-
"loss":
|
| 273 |
"step": 2700
|
| 274 |
},
|
| 275 |
{
|
| 276 |
-
"epoch": 5.
|
| 277 |
-
"grad_norm": 391.66119384765625,
|
| 278 |
"learning_rate": 0.0003,
|
| 279 |
-
"loss":
|
| 280 |
"step": 2800
|
| 281 |
},
|
| 282 |
{
|
| 283 |
-
"epoch": 5.
|
| 284 |
-
"grad_norm": 599371136.0,
|
| 285 |
"learning_rate": 0.0003,
|
| 286 |
-
"loss":
|
| 287 |
"step": 2900
|
| 288 |
},
|
| 289 |
{
|
| 290 |
-
"epoch": 5.
|
| 291 |
-
"grad_norm": 5585.97314453125,
|
| 292 |
"learning_rate": 0.0003,
|
| 293 |
-
"loss":
|
| 294 |
"step": 3000
|
| 295 |
},
|
| 296 |
{
|
| 297 |
-
"epoch": 5.
|
| 298 |
-
"grad_norm": 268239548448768.0,
|
| 299 |
"learning_rate": 0.0003,
|
| 300 |
-
"loss":
|
| 301 |
"step": 3100
|
| 302 |
},
|
| 303 |
{
|
| 304 |
"epoch": 6.0,
|
| 305 |
-
"eval_accuracy": 0.
|
| 306 |
-
"eval_loss":
|
| 307 |
-
"eval_runtime":
|
| 308 |
-
"eval_samples_per_second":
|
| 309 |
-
"eval_steps_per_second":
|
| 310 |
"step": 3174
|
| 311 |
},
|
| 312 |
{
|
|
@@ -316,54 +285,48 @@
|
|
| 316 |
"step": 3174
|
| 317 |
},
|
| 318 |
{
|
| 319 |
-
"epoch": 6.
|
| 320 |
-
"grad_norm": 21535.734375,
|
| 321 |
"learning_rate": 0.0003,
|
| 322 |
-
"loss":
|
| 323 |
"step": 3200
|
| 324 |
},
|
| 325 |
{
|
| 326 |
-
"epoch": 6.
|
| 327 |
-
"grad_norm": 40871530496.0,
|
| 328 |
"learning_rate": 0.0003,
|
| 329 |
-
"loss":
|
| 330 |
"step": 3300
|
| 331 |
},
|
| 332 |
{
|
| 333 |
-
"epoch": 6.
|
| 334 |
-
"grad_norm": 13.242510795593262,
|
| 335 |
"learning_rate": 0.0003,
|
| 336 |
-
"loss":
|
| 337 |
"step": 3400
|
| 338 |
},
|
| 339 |
{
|
| 340 |
-
"epoch": 6.
|
| 341 |
-
"grad_norm": 3.0841150283813477,
|
| 342 |
"learning_rate": 0.0003,
|
| 343 |
-
"loss":
|
| 344 |
"step": 3500
|
| 345 |
},
|
| 346 |
{
|
| 347 |
-
"epoch": 6.
|
| 348 |
-
"grad_norm": 1.6124571561813354,
|
| 349 |
"learning_rate": 0.0003,
|
| 350 |
-
"loss":
|
| 351 |
"step": 3600
|
| 352 |
},
|
| 353 |
{
|
| 354 |
-
"epoch": 6.
|
| 355 |
-
"grad_norm": 1.8358014822006226,
|
| 356 |
"learning_rate": 0.0003,
|
| 357 |
-
"loss":
|
| 358 |
"step": 3700
|
| 359 |
},
|
| 360 |
{
|
| 361 |
"epoch": 7.0,
|
| 362 |
-
"eval_accuracy": 0.
|
| 363 |
-
"eval_loss":
|
| 364 |
-
"eval_runtime":
|
| 365 |
-
"eval_samples_per_second":
|
| 366 |
-
"eval_steps_per_second":
|
| 367 |
"step": 3703
|
| 368 |
},
|
| 369 |
{
|
|
@@ -373,47 +336,42 @@
|
|
| 373 |
"step": 3703
|
| 374 |
},
|
| 375 |
{
|
| 376 |
-
"epoch": 7.
|
| 377 |
-
"grad_norm": 5.786651134490967,
|
| 378 |
"learning_rate": 0.0003,
|
| 379 |
-
"loss":
|
| 380 |
"step": 3800
|
| 381 |
},
|
| 382 |
{
|
| 383 |
-
"epoch": 7.
|
| 384 |
-
"grad_norm": 3.7182490825653076,
|
| 385 |
"learning_rate": 0.0003,
|
| 386 |
-
"loss":
|
| 387 |
"step": 3900
|
| 388 |
},
|
| 389 |
{
|
| 390 |
-
"epoch": 7.
|
| 391 |
-
"grad_norm": 31.241104125976562,
|
| 392 |
"learning_rate": 0.0003,
|
| 393 |
-
"loss": 7.
|
| 394 |
"step": 4000
|
| 395 |
},
|
| 396 |
{
|
| 397 |
-
"epoch": 7.
|
| 398 |
-
"grad_norm": 896688896.0,
|
| 399 |
"learning_rate": 0.0003,
|
| 400 |
-
"loss": 8.
|
| 401 |
"step": 4100
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch": 7.
|
| 405 |
-
"grad_norm": 1.1150538843415904e+19,
|
| 406 |
"learning_rate": 0.0003,
|
| 407 |
-
"loss":
|
| 408 |
"step": 4200
|
| 409 |
},
|
| 410 |
{
|
| 411 |
"epoch": 8.0,
|
| 412 |
-
"eval_accuracy": 0.
|
| 413 |
-
"eval_loss":
|
| 414 |
-
"eval_runtime":
|
| 415 |
-
"eval_samples_per_second":
|
| 416 |
-
"eval_steps_per_second":
|
| 417 |
"step": 4232
|
| 418 |
},
|
| 419 |
{
|
|
@@ -423,47 +381,42 @@
|
|
| 423 |
"step": 4232
|
| 424 |
},
|
| 425 |
{
|
| 426 |
-
"epoch": 8.
|
| 427 |
-
"grad_norm": 7.2513813972473145,
|
| 428 |
"learning_rate": 0.0003,
|
| 429 |
-
"loss":
|
| 430 |
"step": 4300
|
| 431 |
},
|
| 432 |
{
|
| 433 |
-
"epoch": 8.
|
| 434 |
-
"grad_norm": 14.863628387451172,
|
| 435 |
"learning_rate": 0.0003,
|
| 436 |
-
"loss":
|
| 437 |
"step": 4400
|
| 438 |
},
|
| 439 |
{
|
| 440 |
-
"epoch": 8.
|
| 441 |
-
"grad_norm": 5.358551979064941,
|
| 442 |
"learning_rate": 0.0003,
|
| 443 |
-
"loss": 8.
|
| 444 |
"step": 4500
|
| 445 |
},
|
| 446 |
{
|
| 447 |
-
"epoch": 8.
|
| 448 |
-
"grad_norm": 113.71702575683594,
|
| 449 |
"learning_rate": 0.0003,
|
| 450 |
-
"loss": 7.
|
| 451 |
"step": 4600
|
| 452 |
},
|
| 453 |
{
|
| 454 |
-
"epoch": 8.
|
| 455 |
-
"grad_norm": 2.3997488021850586,
|
| 456 |
"learning_rate": 0.0003,
|
| 457 |
-
"loss":
|
| 458 |
"step": 4700
|
| 459 |
},
|
| 460 |
{
|
| 461 |
"epoch": 9.0,
|
| 462 |
-
"eval_accuracy": 0.
|
| 463 |
-
"eval_loss": 6.
|
| 464 |
-
"eval_runtime":
|
| 465 |
-
"eval_samples_per_second":
|
| 466 |
-
"eval_steps_per_second":
|
| 467 |
"step": 4761
|
| 468 |
},
|
| 469 |
{
|
|
@@ -473,47 +426,42 @@
|
|
| 473 |
"step": 4761
|
| 474 |
},
|
| 475 |
{
|
| 476 |
-
"epoch": 9.
|
| 477 |
-
"grad_norm": 7.822358131408691,
|
| 478 |
"learning_rate": 0.0003,
|
| 479 |
-
"loss": 7.
|
| 480 |
"step": 4800
|
| 481 |
},
|
| 482 |
{
|
| 483 |
-
"epoch": 9.
|
| 484 |
-
"grad_norm": 89279359352832.0,
|
| 485 |
"learning_rate": 0.0003,
|
| 486 |
-
"loss": 7.
|
| 487 |
"step": 4900
|
| 488 |
},
|
| 489 |
{
|
| 490 |
-
"epoch": 9.
|
| 491 |
-
"grad_norm": 656.6845703125,
|
| 492 |
"learning_rate": 0.0003,
|
| 493 |
-
"loss": 7.
|
| 494 |
"step": 5000
|
| 495 |
},
|
| 496 |
{
|
| 497 |
-
"epoch": 9.
|
| 498 |
-
"grad_norm": 13841700.0,
|
| 499 |
"learning_rate": 0.0003,
|
| 500 |
-
"loss": 7.
|
| 501 |
"step": 5100
|
| 502 |
},
|
| 503 |
{
|
| 504 |
-
"epoch": 9.
|
| 505 |
-
"grad_norm": 768177.9375,
|
| 506 |
"learning_rate": 0.0003,
|
| 507 |
-
"loss": 7.
|
| 508 |
"step": 5200
|
| 509 |
},
|
| 510 |
{
|
| 511 |
"epoch": 10.0,
|
| 512 |
-
"eval_accuracy": 0.
|
| 513 |
-
"eval_loss":
|
| 514 |
-
"eval_runtime":
|
| 515 |
-
"eval_samples_per_second":
|
| 516 |
-
"eval_steps_per_second":
|
| 517 |
"step": 5290
|
| 518 |
},
|
| 519 |
{
|
|
@@ -523,54 +471,48 @@
|
|
| 523 |
"step": 5290
|
| 524 |
},
|
| 525 |
{
|
| 526 |
-
"epoch": 10.
|
| 527 |
-
"grad_norm": 389293.03125,
|
| 528 |
"learning_rate": 0.0003,
|
| 529 |
-
"loss": 7.
|
| 530 |
"step": 5300
|
| 531 |
},
|
| 532 |
{
|
| 533 |
-
"epoch": 10.
|
| 534 |
-
"grad_norm": 3129429.25,
|
| 535 |
"learning_rate": 0.0003,
|
| 536 |
-
"loss":
|
| 537 |
"step": 5400
|
| 538 |
},
|
| 539 |
{
|
| 540 |
-
"epoch": 10.
|
| 541 |
-
"grad_norm": 6832450.5,
|
| 542 |
"learning_rate": 0.0003,
|
| 543 |
-
"loss":
|
| 544 |
"step": 5500
|
| 545 |
},
|
| 546 |
{
|
| 547 |
-
"epoch": 10.
|
| 548 |
-
"grad_norm": 24994486.0,
|
| 549 |
"learning_rate": 0.0003,
|
| 550 |
-
"loss":
|
| 551 |
"step": 5600
|
| 552 |
},
|
| 553 |
{
|
| 554 |
-
"epoch": 10.
|
| 555 |
-
"grad_norm": 10634.3076171875,
|
| 556 |
"learning_rate": 0.0003,
|
| 557 |
-
"loss":
|
| 558 |
"step": 5700
|
| 559 |
},
|
| 560 |
{
|
| 561 |
-
"epoch": 10.
|
| 562 |
-
"grad_norm": 37.100101470947266,
|
| 563 |
"learning_rate": 0.0003,
|
| 564 |
-
"loss":
|
| 565 |
"step": 5800
|
| 566 |
},
|
| 567 |
{
|
| 568 |
"epoch": 11.0,
|
| 569 |
-
"eval_accuracy": 0.
|
| 570 |
-
"eval_loss": 6.
|
| 571 |
-
"eval_runtime":
|
| 572 |
-
"eval_samples_per_second":
|
| 573 |
-
"eval_steps_per_second":
|
| 574 |
"step": 5819
|
| 575 |
},
|
| 576 |
{
|
|
@@ -580,47 +522,42 @@
|
|
| 580 |
"step": 5819
|
| 581 |
},
|
| 582 |
{
|
| 583 |
-
"epoch": 11.
|
| 584 |
-
"grad_norm": 319.2048645019531,
|
| 585 |
"learning_rate": 0.0003,
|
| 586 |
-
"loss": 7.
|
| 587 |
"step": 5900
|
| 588 |
},
|
| 589 |
{
|
| 590 |
-
"epoch": 11.
|
| 591 |
-
"grad_norm": 72.28472900390625,
|
| 592 |
"learning_rate": 0.0003,
|
| 593 |
-
"loss": 7.
|
| 594 |
"step": 6000
|
| 595 |
},
|
| 596 |
{
|
| 597 |
-
"epoch": 11.
|
| 598 |
-
"grad_norm": 11.851283073425293,
|
| 599 |
"learning_rate": 0.0003,
|
| 600 |
-
"loss": 7.
|
| 601 |
"step": 6100
|
| 602 |
},
|
| 603 |
{
|
| 604 |
-
"epoch": 11.
|
| 605 |
-
"grad_norm": 6.574771404266357,
|
| 606 |
"learning_rate": 0.0003,
|
| 607 |
-
"loss": 7.
|
| 608 |
"step": 6200
|
| 609 |
},
|
| 610 |
{
|
| 611 |
-
"epoch": 11.
|
| 612 |
-
"grad_norm": 4.578986644744873,
|
| 613 |
"learning_rate": 0.0003,
|
| 614 |
-
"loss": 7.
|
| 615 |
"step": 6300
|
| 616 |
},
|
| 617 |
{
|
| 618 |
"epoch": 12.0,
|
| 619 |
-
"eval_accuracy": 0.
|
| 620 |
-
"eval_loss": 6.
|
| 621 |
-
"eval_runtime":
|
| 622 |
-
"eval_samples_per_second":
|
| 623 |
-
"eval_steps_per_second":
|
| 624 |
"step": 6348
|
| 625 |
},
|
| 626 |
{
|
|
@@ -630,47 +567,42 @@
|
|
| 630 |
"step": 6348
|
| 631 |
},
|
| 632 |
{
|
| 633 |
-
"epoch": 12.
|
| 634 |
-
"grad_norm": 21.886821746826172,
|
| 635 |
"learning_rate": 0.0003,
|
| 636 |
-
"loss": 7.
|
| 637 |
"step": 6400
|
| 638 |
},
|
| 639 |
{
|
| 640 |
-
"epoch": 12.
|
| 641 |
-
"grad_norm": 102.8978500366211,
|
| 642 |
"learning_rate": 0.0003,
|
| 643 |
-
"loss": 7.
|
| 644 |
"step": 6500
|
| 645 |
},
|
| 646 |
{
|
| 647 |
-
"epoch": 12.
|
| 648 |
-
"grad_norm": 4.47935676574707,
|
| 649 |
"learning_rate": 0.0003,
|
| 650 |
-
"loss":
|
| 651 |
"step": 6600
|
| 652 |
},
|
| 653 |
{
|
| 654 |
-
"epoch": 12.
|
| 655 |
-
"grad_norm": 259.2627868652344,
|
| 656 |
"learning_rate": 0.0003,
|
| 657 |
-
"loss":
|
| 658 |
"step": 6700
|
| 659 |
},
|
| 660 |
{
|
| 661 |
-
"epoch": 12.
|
| 662 |
-
"grad_norm": 473.829345703125,
|
| 663 |
"learning_rate": 0.0003,
|
| 664 |
-
"loss": 7.
|
| 665 |
"step": 6800
|
| 666 |
},
|
| 667 |
{
|
| 668 |
"epoch": 13.0,
|
| 669 |
-
"eval_accuracy": 0.
|
| 670 |
-
"eval_loss": 6.
|
| 671 |
-
"eval_runtime":
|
| 672 |
-
"eval_samples_per_second":
|
| 673 |
-
"eval_steps_per_second":
|
| 674 |
"step": 6877
|
| 675 |
},
|
| 676 |
{
|
|
@@ -680,54 +612,48 @@
|
|
| 680 |
"step": 6877
|
| 681 |
},
|
| 682 |
{
|
| 683 |
-
"epoch": 13.
|
| 684 |
-
"grad_norm": 150.06381225585938,
|
| 685 |
"learning_rate": 0.0003,
|
| 686 |
-
"loss":
|
| 687 |
"step": 6900
|
| 688 |
},
|
| 689 |
{
|
| 690 |
-
"epoch": 13.
|
| 691 |
-
"grad_norm": 1546.5465087890625,
|
| 692 |
"learning_rate": 0.0003,
|
| 693 |
-
"loss":
|
| 694 |
"step": 7000
|
| 695 |
},
|
| 696 |
{
|
| 697 |
-
"epoch": 13.
|
| 698 |
-
"grad_norm": 18108432.0,
|
| 699 |
"learning_rate": 0.0003,
|
| 700 |
-
"loss": 7.
|
| 701 |
"step": 7100
|
| 702 |
},
|
| 703 |
{
|
| 704 |
-
"epoch": 13.
|
| 705 |
-
"grad_norm": 379725545472.0,
|
| 706 |
"learning_rate": 0.0003,
|
| 707 |
-
"loss":
|
| 708 |
"step": 7200
|
| 709 |
},
|
| 710 |
{
|
| 711 |
-
"epoch": 13.
|
| 712 |
-
"grad_norm": 14.351493835449219,
|
| 713 |
"learning_rate": 0.0003,
|
| 714 |
-
"loss": 7.
|
| 715 |
"step": 7300
|
| 716 |
},
|
| 717 |
{
|
| 718 |
-
"epoch": 13.
|
| 719 |
-
"grad_norm": 3.4325509071350098,
|
| 720 |
"learning_rate": 0.0003,
|
| 721 |
-
"loss":
|
| 722 |
"step": 7400
|
| 723 |
},
|
| 724 |
{
|
| 725 |
"epoch": 14.0,
|
| 726 |
-
"eval_accuracy": 0.
|
| 727 |
-
"eval_loss": 6.
|
| 728 |
-
"eval_runtime":
|
| 729 |
-
"eval_samples_per_second":
|
| 730 |
-
"eval_steps_per_second":
|
| 731 |
"step": 7406
|
| 732 |
},
|
| 733 |
{
|
|
@@ -737,47 +663,42 @@
|
|
| 737 |
"step": 7406
|
| 738 |
},
|
| 739 |
{
|
| 740 |
-
"epoch": 14.
|
| 741 |
-
"grad_norm": 3408474406912.0,
|
| 742 |
"learning_rate": 0.0003,
|
| 743 |
-
"loss": 7.
|
| 744 |
"step": 7500
|
| 745 |
},
|
| 746 |
{
|
| 747 |
-
"epoch": 14.
|
| 748 |
-
"grad_norm": 766890.1875,
|
| 749 |
"learning_rate": 0.0003,
|
| 750 |
-
"loss": 7.
|
| 751 |
"step": 7600
|
| 752 |
},
|
| 753 |
{
|
| 754 |
-
"epoch": 14.
|
| 755 |
-
"grad_norm": 4749101.0,
|
| 756 |
"learning_rate": 0.0003,
|
| 757 |
-
"loss": 7.
|
| 758 |
"step": 7700
|
| 759 |
},
|
| 760 |
{
|
| 761 |
-
"epoch": 14.
|
| 762 |
-
"grad_norm": 21509612.0,
|
| 763 |
"learning_rate": 0.0003,
|
| 764 |
-
"loss": 7.
|
| 765 |
"step": 7800
|
| 766 |
},
|
| 767 |
{
|
| 768 |
-
"epoch": 14.
|
| 769 |
-
"grad_norm": 1.0536716136244838e+18,
|
| 770 |
"learning_rate": 0.0003,
|
| 771 |
-
"loss": 7.
|
| 772 |
"step": 7900
|
| 773 |
},
|
| 774 |
{
|
| 775 |
"epoch": 15.0,
|
| 776 |
-
"eval_accuracy": 0.
|
| 777 |
-
"eval_loss":
|
| 778 |
-
"eval_runtime":
|
| 779 |
-
"eval_samples_per_second":
|
| 780 |
-
"eval_steps_per_second":
|
| 781 |
"step": 7935
|
| 782 |
},
|
| 783 |
{
|
|
@@ -787,47 +708,42 @@
|
|
| 787 |
"step": 7935
|
| 788 |
},
|
| 789 |
{
|
| 790 |
-
"epoch": 15.
|
| 791 |
-
"grad_norm": NaN,
|
| 792 |
"learning_rate": 0.0003,
|
| 793 |
-
"loss":
|
| 794 |
"step": 8000
|
| 795 |
},
|
| 796 |
{
|
| 797 |
-
"epoch": 15.
|
| 798 |
-
"grad_norm": NaN,
|
| 799 |
"learning_rate": 0.0003,
|
| 800 |
-
"loss":
|
| 801 |
"step": 8100
|
| 802 |
},
|
| 803 |
{
|
| 804 |
-
"epoch": 15.
|
| 805 |
-
"grad_norm": NaN,
|
| 806 |
"learning_rate": 0.0003,
|
| 807 |
-
"loss":
|
| 808 |
"step": 8200
|
| 809 |
},
|
| 810 |
{
|
| 811 |
-
"epoch": 15.
|
| 812 |
-
"grad_norm": NaN,
|
| 813 |
"learning_rate": 0.0003,
|
| 814 |
-
"loss":
|
| 815 |
"step": 8300
|
| 816 |
},
|
| 817 |
{
|
| 818 |
-
"epoch": 15.
|
| 819 |
-
"grad_norm": NaN,
|
| 820 |
"learning_rate": 0.0003,
|
| 821 |
-
"loss":
|
| 822 |
"step": 8400
|
| 823 |
},
|
| 824 |
{
|
| 825 |
"epoch": 16.0,
|
| 826 |
-
"eval_accuracy": 0.
|
| 827 |
-
"eval_loss":
|
| 828 |
-
"eval_runtime":
|
| 829 |
-
"eval_samples_per_second":
|
| 830 |
-
"eval_steps_per_second":
|
| 831 |
"step": 8464
|
| 832 |
},
|
| 833 |
{
|
|
@@ -837,47 +753,42 @@
|
|
| 837 |
"step": 8464
|
| 838 |
},
|
| 839 |
{
|
| 840 |
-
"epoch": 16.
|
| 841 |
-
"grad_norm": NaN,
|
| 842 |
"learning_rate": 0.0003,
|
| 843 |
-
"loss":
|
| 844 |
"step": 8500
|
| 845 |
},
|
| 846 |
{
|
| 847 |
-
"epoch": 16.
|
| 848 |
-
"grad_norm": NaN,
|
| 849 |
"learning_rate": 0.0003,
|
| 850 |
-
"loss":
|
| 851 |
"step": 8600
|
| 852 |
},
|
| 853 |
{
|
| 854 |
-
"epoch": 16.
|
| 855 |
-
"grad_norm": NaN,
|
| 856 |
"learning_rate": 0.0003,
|
| 857 |
-
"loss":
|
| 858 |
"step": 8700
|
| 859 |
},
|
| 860 |
{
|
| 861 |
-
"epoch": 16.
|
| 862 |
-
"grad_norm": NaN,
|
| 863 |
"learning_rate": 0.0003,
|
| 864 |
-
"loss":
|
| 865 |
"step": 8800
|
| 866 |
},
|
| 867 |
{
|
| 868 |
-
"epoch": 16.
|
| 869 |
-
"grad_norm": NaN,
|
| 870 |
"learning_rate": 0.0003,
|
| 871 |
-
"loss":
|
| 872 |
"step": 8900
|
| 873 |
},
|
| 874 |
{
|
| 875 |
"epoch": 17.0,
|
| 876 |
-
"eval_accuracy": 0.
|
| 877 |
-
"eval_loss":
|
| 878 |
-
"eval_runtime":
|
| 879 |
-
"eval_samples_per_second":
|
| 880 |
-
"eval_steps_per_second":
|
| 881 |
"step": 8993
|
| 882 |
},
|
| 883 |
{
|
|
@@ -887,54 +798,48 @@
|
|
| 887 |
"step": 8993
|
| 888 |
},
|
| 889 |
{
|
| 890 |
-
"epoch": 17.
|
| 891 |
-
"grad_norm": NaN,
|
| 892 |
"learning_rate": 0.0003,
|
| 893 |
-
"loss":
|
| 894 |
"step": 9000
|
| 895 |
},
|
| 896 |
{
|
| 897 |
-
"epoch": 17.
|
| 898 |
-
"grad_norm": NaN,
|
| 899 |
"learning_rate": 0.0003,
|
| 900 |
-
"loss":
|
| 901 |
"step": 9100
|
| 902 |
},
|
| 903 |
{
|
| 904 |
-
"epoch": 17.
|
| 905 |
-
"grad_norm": NaN,
|
| 906 |
"learning_rate": 0.0003,
|
| 907 |
-
"loss":
|
| 908 |
"step": 9200
|
| 909 |
},
|
| 910 |
{
|
| 911 |
-
"epoch": 17.
|
| 912 |
-
"grad_norm": NaN,
|
| 913 |
"learning_rate": 0.0003,
|
| 914 |
-
"loss":
|
| 915 |
"step": 9300
|
| 916 |
},
|
| 917 |
{
|
| 918 |
-
"epoch": 17.
|
| 919 |
-
"grad_norm": NaN,
|
| 920 |
"learning_rate": 0.0003,
|
| 921 |
-
"loss":
|
| 922 |
"step": 9400
|
| 923 |
},
|
| 924 |
{
|
| 925 |
-
"epoch": 17.
|
| 926 |
-
"grad_norm": NaN,
|
| 927 |
"learning_rate": 0.0003,
|
| 928 |
-
"loss":
|
| 929 |
"step": 9500
|
| 930 |
},
|
| 931 |
{
|
| 932 |
"epoch": 18.0,
|
| 933 |
-
"eval_accuracy": 0.
|
| 934 |
-
"eval_loss":
|
| 935 |
-
"eval_runtime":
|
| 936 |
-
"eval_samples_per_second":
|
| 937 |
-
"eval_steps_per_second":
|
| 938 |
"step": 9522
|
| 939 |
},
|
| 940 |
{
|
|
@@ -944,47 +849,42 @@
|
|
| 944 |
"step": 9522
|
| 945 |
},
|
| 946 |
{
|
| 947 |
-
"epoch": 18.
|
| 948 |
-
"grad_norm": NaN,
|
| 949 |
"learning_rate": 0.0003,
|
| 950 |
-
"loss":
|
| 951 |
"step": 9600
|
| 952 |
},
|
| 953 |
{
|
| 954 |
-
"epoch": 18.
|
| 955 |
-
"grad_norm": NaN,
|
| 956 |
"learning_rate": 0.0003,
|
| 957 |
-
"loss":
|
| 958 |
"step": 9700
|
| 959 |
},
|
| 960 |
{
|
| 961 |
-
"epoch": 18.
|
| 962 |
-
"grad_norm": NaN,
|
| 963 |
"learning_rate": 0.0003,
|
| 964 |
-
"loss":
|
| 965 |
"step": 9800
|
| 966 |
},
|
| 967 |
{
|
| 968 |
-
"epoch": 18.
|
| 969 |
-
"grad_norm": NaN,
|
| 970 |
"learning_rate": 0.0003,
|
| 971 |
-
"loss":
|
| 972 |
"step": 9900
|
| 973 |
},
|
| 974 |
{
|
| 975 |
-
"epoch": 18.
|
| 976 |
-
"grad_norm": NaN,
|
| 977 |
"learning_rate": 0.0003,
|
| 978 |
-
"loss":
|
| 979 |
"step": 10000
|
| 980 |
},
|
| 981 |
{
|
| 982 |
"epoch": 19.0,
|
| 983 |
-
"eval_accuracy": 0.
|
| 984 |
-
"eval_loss":
|
| 985 |
-
"eval_runtime":
|
| 986 |
-
"eval_samples_per_second":
|
| 987 |
-
"eval_steps_per_second":
|
| 988 |
"step": 10051
|
| 989 |
},
|
| 990 |
{
|
|
@@ -994,47 +894,42 @@
|
|
| 994 |
"step": 10051
|
| 995 |
},
|
| 996 |
{
|
| 997 |
-
"epoch": 19.
|
| 998 |
-
"grad_norm": NaN,
|
| 999 |
"learning_rate": 0.0003,
|
| 1000 |
-
"loss":
|
| 1001 |
"step": 10100
|
| 1002 |
},
|
| 1003 |
{
|
| 1004 |
-
"epoch": 19.
|
| 1005 |
-
"grad_norm": NaN,
|
| 1006 |
"learning_rate": 0.0003,
|
| 1007 |
-
"loss":
|
| 1008 |
"step": 10200
|
| 1009 |
},
|
| 1010 |
{
|
| 1011 |
-
"epoch": 19.
|
| 1012 |
-
"grad_norm": NaN,
|
| 1013 |
"learning_rate": 0.0003,
|
| 1014 |
-
"loss":
|
| 1015 |
"step": 10300
|
| 1016 |
},
|
| 1017 |
{
|
| 1018 |
-
"epoch": 19.
|
| 1019 |
-
"grad_norm": NaN,
|
| 1020 |
"learning_rate": 0.0003,
|
| 1021 |
-
"loss":
|
| 1022 |
"step": 10400
|
| 1023 |
},
|
| 1024 |
{
|
| 1025 |
-
"epoch": 19.
|
| 1026 |
-
"grad_norm": NaN,
|
| 1027 |
"learning_rate": 0.0003,
|
| 1028 |
-
"loss":
|
| 1029 |
"step": 10500
|
| 1030 |
},
|
| 1031 |
{
|
| 1032 |
"epoch": 20.0,
|
| 1033 |
-
"eval_accuracy": 0.
|
| 1034 |
-
"eval_loss":
|
| 1035 |
-
"eval_runtime":
|
| 1036 |
-
"eval_samples_per_second":
|
| 1037 |
-
"eval_steps_per_second":
|
| 1038 |
"step": 10580
|
| 1039 |
},
|
| 1040 |
{
|
|
@@ -1046,23 +941,9 @@
|
|
| 1046 |
],
|
| 1047 |
"logging_steps": 100,
|
| 1048 |
"max_steps": 26450,
|
| 1049 |
-
"num_input_tokens_seen": 0,
|
| 1050 |
"num_train_epochs": 50,
|
| 1051 |
"save_steps": 500,
|
| 1052 |
-
"
|
| 1053 |
-
"TrainerControl": {
|
| 1054 |
-
"args": {
|
| 1055 |
-
"should_epoch_stop": false,
|
| 1056 |
-
"should_evaluate": false,
|
| 1057 |
-
"should_log": false,
|
| 1058 |
-
"should_save": true,
|
| 1059 |
-
"should_training_stop": false
|
| 1060 |
-
},
|
| 1061 |
-
"attributes": {}
|
| 1062 |
-
}
|
| 1063 |
-
},
|
| 1064 |
-
"total_flos": 8.612663832063508e+17,
|
| 1065 |
-
"train_batch_size": 2,
|
| 1066 |
"trial_name": null,
|
| 1067 |
"trial_params": null
|
| 1068 |
}
|
|
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.19,
|
|
|
|
| 13 |
"learning_rate": 0.0003,
|
| 14 |
+
"loss": 1.4758,
|
| 15 |
"step": 100
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"epoch": 0.38,
|
|
|
|
| 19 |
"learning_rate": 0.0003,
|
| 20 |
+
"loss": 1.3867,
|
| 21 |
"step": 200
|
| 22 |
},
|
| 23 |
{
|
| 24 |
+
"epoch": 0.57,
|
|
|
|
| 25 |
"learning_rate": 0.0003,
|
| 26 |
+
"loss": 1.4064,
|
| 27 |
"step": 300
|
| 28 |
},
|
| 29 |
{
|
| 30 |
+
"epoch": 0.76,
|
|
|
|
| 31 |
"learning_rate": 0.0003,
|
| 32 |
+
"loss": 1.3789,
|
| 33 |
"step": 400
|
| 34 |
},
|
| 35 |
{
|
| 36 |
+
"epoch": 0.95,
|
|
|
|
| 37 |
"learning_rate": 0.0003,
|
| 38 |
+
"loss": 1.3822,
|
| 39 |
"step": 500
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 1.0,
|
| 43 |
+
"eval_accuracy": 0.6171794871794872,
|
| 44 |
+
"eval_loss": 1.2976911067962646,
|
| 45 |
+
"eval_runtime": 4.224,
|
| 46 |
+
"eval_samples_per_second": 118.372,
|
| 47 |
+
"eval_steps_per_second": 14.915,
|
| 48 |
"step": 529
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.0,
|
| 52 |
+
"eval_exact_match": 0.2,
|
| 53 |
+
"eval_f1": 2.031878046337681,
|
| 54 |
"step": 529
|
| 55 |
},
|
| 56 |
{
|
| 57 |
+
"epoch": 1.13,
|
|
|
|
| 58 |
"learning_rate": 0.0003,
|
| 59 |
+
"loss": 1.3048,
|
| 60 |
"step": 600
|
| 61 |
},
|
| 62 |
{
|
| 63 |
+
"epoch": 1.32,
|
|
|
|
| 64 |
"learning_rate": 0.0003,
|
| 65 |
+
"loss": 1.2524,
|
| 66 |
"step": 700
|
| 67 |
},
|
| 68 |
{
|
| 69 |
+
"epoch": 1.51,
|
|
|
|
| 70 |
"learning_rate": 0.0003,
|
| 71 |
+
"loss": 1.3009,
|
| 72 |
"step": 800
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 1.7,
|
|
|
|
| 76 |
"learning_rate": 0.0003,
|
| 77 |
+
"loss": 1.268,
|
| 78 |
"step": 900
|
| 79 |
},
|
| 80 |
{
|
| 81 |
+
"epoch": 1.89,
|
|
|
|
| 82 |
"learning_rate": 0.0003,
|
| 83 |
+
"loss": 1.2744,
|
| 84 |
"step": 1000
|
| 85 |
},
|
| 86 |
{
|
| 87 |
"epoch": 2.0,
|
| 88 |
+
"eval_accuracy": 0.6032307692307692,
|
| 89 |
+
"eval_loss": 1.374481201171875,
|
| 90 |
+
"eval_runtime": 4.4141,
|
| 91 |
+
"eval_samples_per_second": 113.274,
|
| 92 |
+
"eval_steps_per_second": 14.272,
|
| 93 |
"step": 1058
|
| 94 |
},
|
| 95 |
{
|
|
|
|
| 99 |
"step": 1058
|
| 100 |
},
|
| 101 |
{
|
| 102 |
+
"epoch": 2.08,
|
|
|
|
| 103 |
"learning_rate": 0.0003,
|
| 104 |
+
"loss": 1.2158,
|
| 105 |
"step": 1100
|
| 106 |
},
|
| 107 |
{
|
| 108 |
+
"epoch": 2.27,
|
|
|
|
| 109 |
"learning_rate": 0.0003,
|
| 110 |
+
"loss": 1.1066,
|
| 111 |
"step": 1200
|
| 112 |
},
|
| 113 |
{
|
| 114 |
+
"epoch": 2.46,
|
|
|
|
| 115 |
"learning_rate": 0.0003,
|
| 116 |
+
"loss": 1.1365,
|
| 117 |
"step": 1300
|
| 118 |
},
|
| 119 |
{
|
| 120 |
+
"epoch": 2.65,
|
|
|
|
| 121 |
"learning_rate": 0.0003,
|
| 122 |
+
"loss": 1.1486,
|
| 123 |
"step": 1400
|
| 124 |
},
|
| 125 |
{
|
| 126 |
+
"epoch": 2.84,
|
|
|
|
| 127 |
"learning_rate": 0.0003,
|
| 128 |
+
"loss": 1.1768,
|
| 129 |
"step": 1500
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 3.0,
|
| 133 |
+
"eval_accuracy": 0.6157435897435898,
|
| 134 |
+
"eval_loss": 1.331903100013733,
|
| 135 |
+
"eval_runtime": 4.8466,
|
| 136 |
+
"eval_samples_per_second": 103.164,
|
| 137 |
+
"eval_steps_per_second": 12.999,
|
| 138 |
"step": 1587
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 3.0,
|
| 142 |
"eval_exact_match": 0.0,
|
| 143 |
+
"eval_f1": 0.0,
|
| 144 |
"step": 1587
|
| 145 |
},
|
| 146 |
{
|
| 147 |
+
"epoch": 3.02,
|
|
|
|
| 148 |
"learning_rate": 0.0003,
|
| 149 |
+
"loss": 1.1769,
|
| 150 |
"step": 1600
|
| 151 |
},
|
| 152 |
{
|
| 153 |
+
"epoch": 3.21,
|
|
|
|
| 154 |
"learning_rate": 0.0003,
|
| 155 |
+
"loss": 0.9541,
|
| 156 |
"step": 1700
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 3.4,
|
|
|
|
| 160 |
"learning_rate": 0.0003,
|
| 161 |
+
"loss": 1.2939,
|
| 162 |
"step": 1800
|
| 163 |
},
|
| 164 |
{
|
| 165 |
+
"epoch": 3.59,
|
|
|
|
| 166 |
"learning_rate": 0.0003,
|
| 167 |
+
"loss": 0.9852,
|
| 168 |
"step": 1900
|
| 169 |
},
|
| 170 |
{
|
| 171 |
+
"epoch": 3.78,
|
|
|
|
| 172 |
"learning_rate": 0.0003,
|
| 173 |
+
"loss": 0.9538,
|
| 174 |
"step": 2000
|
| 175 |
},
|
| 176 |
{
|
| 177 |
+
"epoch": 3.97,
|
|
|
|
| 178 |
"learning_rate": 0.0003,
|
| 179 |
+
"loss": 0.9247,
|
| 180 |
"step": 2100
|
| 181 |
},
|
| 182 |
{
|
| 183 |
"epoch": 4.0,
|
| 184 |
+
"eval_accuracy": 0.6101538461538462,
|
| 185 |
+
"eval_loss": 1.4367107152938843,
|
| 186 |
+
"eval_runtime": 4.4146,
|
| 187 |
+
"eval_samples_per_second": 113.259,
|
| 188 |
+
"eval_steps_per_second": 14.271,
|
| 189 |
"step": 2116
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"epoch": 4.0,
|
| 193 |
+
"eval_exact_match": 6.6,
|
| 194 |
+
"eval_f1": 9.897142857142855,
|
| 195 |
"step": 2116
|
| 196 |
},
|
| 197 |
{
|
| 198 |
+
"epoch": 4.16,
|
|
|
|
| 199 |
"learning_rate": 0.0003,
|
| 200 |
+
"loss": 1.1334,
|
| 201 |
"step": 2200
|
| 202 |
},
|
| 203 |
{
|
| 204 |
+
"epoch": 4.35,
|
|
|
|
| 205 |
"learning_rate": 0.0003,
|
| 206 |
+
"loss": 1.5175,
|
| 207 |
"step": 2300
|
| 208 |
},
|
| 209 |
{
|
| 210 |
+
"epoch": 4.54,
|
|
|
|
| 211 |
"learning_rate": 0.0003,
|
| 212 |
+
"loss": 4.2745,
|
| 213 |
"step": 2400
|
| 214 |
},
|
| 215 |
{
|
| 216 |
+
"epoch": 4.73,
|
|
|
|
| 217 |
"learning_rate": 0.0003,
|
| 218 |
+
"loss": 2.5667,
|
| 219 |
"step": 2500
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 4.91,
|
|
|
|
| 223 |
"learning_rate": 0.0003,
|
| 224 |
+
"loss": 1.1836,
|
| 225 |
"step": 2600
|
| 226 |
},
|
| 227 |
{
|
| 228 |
"epoch": 5.0,
|
| 229 |
+
"eval_accuracy": 0.556923076923077,
|
| 230 |
+
"eval_loss": 1.9167817831039429,
|
| 231 |
+
"eval_runtime": 4.8432,
|
| 232 |
+
"eval_samples_per_second": 103.238,
|
| 233 |
+
"eval_steps_per_second": 13.008,
|
| 234 |
"step": 2645
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 5.0,
|
| 238 |
"eval_exact_match": 0.0,
|
| 239 |
+
"eval_f1": 0.2264729350912329,
|
| 240 |
"step": 2645
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 5.1,
|
|
|
|
| 244 |
"learning_rate": 0.0003,
|
| 245 |
+
"loss": 1.7226,
|
| 246 |
"step": 2700
|
| 247 |
},
|
| 248 |
{
|
| 249 |
+
"epoch": 5.29,
|
|
|
|
| 250 |
"learning_rate": 0.0003,
|
| 251 |
+
"loss": 2.1466,
|
| 252 |
"step": 2800
|
| 253 |
},
|
| 254 |
{
|
| 255 |
+
"epoch": 5.48,
|
|
|
|
| 256 |
"learning_rate": 0.0003,
|
| 257 |
+
"loss": 1.83,
|
| 258 |
"step": 2900
|
| 259 |
},
|
| 260 |
{
|
| 261 |
+
"epoch": 5.67,
|
|
|
|
| 262 |
"learning_rate": 0.0003,
|
| 263 |
+
"loss": 1.8789,
|
| 264 |
"step": 3000
|
| 265 |
},
|
| 266 |
{
|
| 267 |
+
"epoch": 5.86,
|
|
|
|
| 268 |
"learning_rate": 0.0003,
|
| 269 |
+
"loss": 2.035,
|
| 270 |
"step": 3100
|
| 271 |
},
|
| 272 |
{
|
| 273 |
"epoch": 6.0,
|
| 274 |
+
"eval_accuracy": 0.5377435897435897,
|
| 275 |
+
"eval_loss": 2.0794410705566406,
|
| 276 |
+
"eval_runtime": 4.1284,
|
| 277 |
+
"eval_samples_per_second": 121.112,
|
| 278 |
+
"eval_steps_per_second": 15.26,
|
| 279 |
"step": 3174
|
| 280 |
},
|
| 281 |
{
|
|
|
|
| 285 |
"step": 3174
|
| 286 |
},
|
| 287 |
{
|
| 288 |
+
"epoch": 6.05,
|
|
|
|
| 289 |
"learning_rate": 0.0003,
|
| 290 |
+
"loss": 2.2251,
|
| 291 |
"step": 3200
|
| 292 |
},
|
| 293 |
{
|
| 294 |
+
"epoch": 6.24,
|
|
|
|
| 295 |
"learning_rate": 0.0003,
|
| 296 |
+
"loss": 2.1457,
|
| 297 |
"step": 3300
|
| 298 |
},
|
| 299 |
{
|
| 300 |
+
"epoch": 6.43,
|
|
|
|
| 301 |
"learning_rate": 0.0003,
|
| 302 |
+
"loss": 2.0764,
|
| 303 |
"step": 3400
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 6.62,
|
|
|
|
| 307 |
"learning_rate": 0.0003,
|
| 308 |
+
"loss": 2.3695,
|
| 309 |
"step": 3500
|
| 310 |
},
|
| 311 |
{
|
| 312 |
+
"epoch": 6.81,
|
|
|
|
| 313 |
"learning_rate": 0.0003,
|
| 314 |
+
"loss": 2.2254,
|
| 315 |
"step": 3600
|
| 316 |
},
|
| 317 |
{
|
| 318 |
+
"epoch": 6.99,
|
|
|
|
| 319 |
"learning_rate": 0.0003,
|
| 320 |
+
"loss": 3.7483,
|
| 321 |
"step": 3700
|
| 322 |
},
|
| 323 |
{
|
| 324 |
"epoch": 7.0,
|
| 325 |
+
"eval_accuracy": 0.48805128205128206,
|
| 326 |
+
"eval_loss": 2.6722612380981445,
|
| 327 |
+
"eval_runtime": 4.2095,
|
| 328 |
+
"eval_samples_per_second": 118.778,
|
| 329 |
+
"eval_steps_per_second": 14.966,
|
| 330 |
"step": 3703
|
| 331 |
},
|
| 332 |
{
|
|
|
|
| 336 |
"step": 3703
|
| 337 |
},
|
| 338 |
{
|
| 339 |
+
"epoch": 7.18,
|
|
|
|
| 340 |
"learning_rate": 0.0003,
|
| 341 |
+
"loss": 6.5533,
|
| 342 |
"step": 3800
|
| 343 |
},
|
| 344 |
{
|
| 345 |
+
"epoch": 7.37,
|
|
|
|
| 346 |
"learning_rate": 0.0003,
|
| 347 |
+
"loss": 6.708,
|
| 348 |
"step": 3900
|
| 349 |
},
|
| 350 |
{
|
| 351 |
+
"epoch": 7.56,
|
|
|
|
| 352 |
"learning_rate": 0.0003,
|
| 353 |
+
"loss": 7.4301,
|
| 354 |
"step": 4000
|
| 355 |
},
|
| 356 |
{
|
| 357 |
+
"epoch": 7.75,
|
|
|
|
| 358 |
"learning_rate": 0.0003,
|
| 359 |
+
"loss": 8.6731,
|
| 360 |
"step": 4100
|
| 361 |
},
|
| 362 |
{
|
| 363 |
+
"epoch": 7.94,
|
|
|
|
| 364 |
"learning_rate": 0.0003,
|
| 365 |
+
"loss": 7.127,
|
| 366 |
"step": 4200
|
| 367 |
},
|
| 368 |
{
|
| 369 |
"epoch": 8.0,
|
| 370 |
+
"eval_accuracy": 0.1922051282051282,
|
| 371 |
+
"eval_loss": 7.040984153747559,
|
| 372 |
+
"eval_runtime": 4.8262,
|
| 373 |
+
"eval_samples_per_second": 103.6,
|
| 374 |
+
"eval_steps_per_second": 13.054,
|
| 375 |
"step": 4232
|
| 376 |
},
|
| 377 |
{
|
|
|
|
| 381 |
"step": 4232
|
| 382 |
},
|
| 383 |
{
|
| 384 |
+
"epoch": 8.13,
|
|
|
|
| 385 |
"learning_rate": 0.0003,
|
| 386 |
+
"loss": 7.8796,
|
| 387 |
"step": 4300
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 8.32,
|
|
|
|
| 391 |
"learning_rate": 0.0003,
|
| 392 |
+
"loss": 8.0716,
|
| 393 |
"step": 4400
|
| 394 |
},
|
| 395 |
{
|
| 396 |
+
"epoch": 8.51,
|
|
|
|
| 397 |
"learning_rate": 0.0003,
|
| 398 |
+
"loss": 8.2395,
|
| 399 |
"step": 4500
|
| 400 |
},
|
| 401 |
{
|
| 402 |
+
"epoch": 8.7,
|
|
|
|
| 403 |
"learning_rate": 0.0003,
|
| 404 |
+
"loss": 7.9157,
|
| 405 |
"step": 4600
|
| 406 |
},
|
| 407 |
{
|
| 408 |
+
"epoch": 8.88,
|
|
|
|
| 409 |
"learning_rate": 0.0003,
|
| 410 |
+
"loss": 7.5321,
|
| 411 |
"step": 4700
|
| 412 |
},
|
| 413 |
{
|
| 414 |
"epoch": 9.0,
|
| 415 |
+
"eval_accuracy": 0.19405128205128205,
|
| 416 |
+
"eval_loss": 6.64878511428833,
|
| 417 |
+
"eval_runtime": 4.104,
|
| 418 |
+
"eval_samples_per_second": 121.831,
|
| 419 |
+
"eval_steps_per_second": 15.351,
|
| 420 |
"step": 4761
|
| 421 |
},
|
| 422 |
{
|
|
|
|
| 426 |
"step": 4761
|
| 427 |
},
|
| 428 |
{
|
| 429 |
+
"epoch": 9.07,
|
|
|
|
| 430 |
"learning_rate": 0.0003,
|
| 431 |
+
"loss": 7.2432,
|
| 432 |
"step": 4800
|
| 433 |
},
|
| 434 |
{
|
| 435 |
+
"epoch": 9.26,
|
|
|
|
| 436 |
"learning_rate": 0.0003,
|
| 437 |
+
"loss": 7.4268,
|
| 438 |
"step": 4900
|
| 439 |
},
|
| 440 |
{
|
| 441 |
+
"epoch": 9.45,
|
|
|
|
| 442 |
"learning_rate": 0.0003,
|
| 443 |
+
"loss": 7.7517,
|
| 444 |
"step": 5000
|
| 445 |
},
|
| 446 |
{
|
| 447 |
+
"epoch": 9.64,
|
|
|
|
| 448 |
"learning_rate": 0.0003,
|
| 449 |
+
"loss": 7.4674,
|
| 450 |
"step": 5100
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"epoch": 9.83,
|
|
|
|
| 454 |
"learning_rate": 0.0003,
|
| 455 |
+
"loss": 7.3806,
|
| 456 |
"step": 5200
|
| 457 |
},
|
| 458 |
{
|
| 459 |
"epoch": 10.0,
|
| 460 |
+
"eval_accuracy": 0.2196923076923077,
|
| 461 |
+
"eval_loss": 6.842738151550293,
|
| 462 |
+
"eval_runtime": 4.8429,
|
| 463 |
+
"eval_samples_per_second": 103.244,
|
| 464 |
+
"eval_steps_per_second": 13.009,
|
| 465 |
"step": 5290
|
| 466 |
},
|
| 467 |
{
|
|
|
|
| 471 |
"step": 5290
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 10.02,
|
|
|
|
| 475 |
"learning_rate": 0.0003,
|
| 476 |
+
"loss": 7.4491,
|
| 477 |
"step": 5300
|
| 478 |
},
|
| 479 |
{
|
| 480 |
+
"epoch": 10.21,
|
|
|
|
| 481 |
"learning_rate": 0.0003,
|
| 482 |
+
"loss": 7.4851,
|
| 483 |
"step": 5400
|
| 484 |
},
|
| 485 |
{
|
| 486 |
+
"epoch": 10.4,
|
|
|
|
| 487 |
"learning_rate": 0.0003,
|
| 488 |
+
"loss": 7.5747,
|
| 489 |
"step": 5500
|
| 490 |
},
|
| 491 |
{
|
| 492 |
+
"epoch": 10.59,
|
|
|
|
| 493 |
"learning_rate": 0.0003,
|
| 494 |
+
"loss": 7.7881,
|
| 495 |
"step": 5600
|
| 496 |
},
|
| 497 |
{
|
| 498 |
+
"epoch": 10.78,
|
|
|
|
| 499 |
"learning_rate": 0.0003,
|
| 500 |
+
"loss": 7.8095,
|
| 501 |
"step": 5700
|
| 502 |
},
|
| 503 |
{
|
| 504 |
+
"epoch": 10.96,
|
|
|
|
| 505 |
"learning_rate": 0.0003,
|
| 506 |
+
"loss": 7.8159,
|
| 507 |
"step": 5800
|
| 508 |
},
|
| 509 |
{
|
| 510 |
"epoch": 11.0,
|
| 511 |
+
"eval_accuracy": 0.2196923076923077,
|
| 512 |
+
"eval_loss": 6.883613586425781,
|
| 513 |
+
"eval_runtime": 4.1075,
|
| 514 |
+
"eval_samples_per_second": 121.727,
|
| 515 |
+
"eval_steps_per_second": 15.338,
|
| 516 |
"step": 5819
|
| 517 |
},
|
| 518 |
{
|
|
|
|
| 522 |
"step": 5819
|
| 523 |
},
|
| 524 |
{
|
| 525 |
+
"epoch": 11.15,
|
|
|
|
| 526 |
"learning_rate": 0.0003,
|
| 527 |
+
"loss": 7.7222,
|
| 528 |
"step": 5900
|
| 529 |
},
|
| 530 |
{
|
| 531 |
+
"epoch": 11.34,
|
|
|
|
| 532 |
"learning_rate": 0.0003,
|
| 533 |
+
"loss": 7.6203,
|
| 534 |
"step": 6000
|
| 535 |
},
|
| 536 |
{
|
| 537 |
+
"epoch": 11.53,
|
|
|
|
| 538 |
"learning_rate": 0.0003,
|
| 539 |
+
"loss": 7.8391,
|
| 540 |
"step": 6100
|
| 541 |
},
|
| 542 |
{
|
| 543 |
+
"epoch": 11.72,
|
|
|
|
| 544 |
"learning_rate": 0.0003,
|
| 545 |
+
"loss": 7.8888,
|
| 546 |
"step": 6200
|
| 547 |
},
|
| 548 |
{
|
| 549 |
+
"epoch": 11.91,
|
|
|
|
| 550 |
"learning_rate": 0.0003,
|
| 551 |
+
"loss": 7.975,
|
| 552 |
"step": 6300
|
| 553 |
},
|
| 554 |
{
|
| 555 |
"epoch": 12.0,
|
| 556 |
+
"eval_accuracy": 0.2196923076923077,
|
| 557 |
+
"eval_loss": 6.876262187957764,
|
| 558 |
+
"eval_runtime": 4.4184,
|
| 559 |
+
"eval_samples_per_second": 113.162,
|
| 560 |
+
"eval_steps_per_second": 14.258,
|
| 561 |
"step": 6348
|
| 562 |
},
|
| 563 |
{
|
|
|
|
| 567 |
"step": 6348
|
| 568 |
},
|
| 569 |
{
|
| 570 |
+
"epoch": 12.1,
|
|
|
|
| 571 |
"learning_rate": 0.0003,
|
| 572 |
+
"loss": 7.9981,
|
| 573 |
"step": 6400
|
| 574 |
},
|
| 575 |
{
|
| 576 |
+
"epoch": 12.29,
|
|
|
|
| 577 |
"learning_rate": 0.0003,
|
| 578 |
+
"loss": 7.9944,
|
| 579 |
"step": 6500
|
| 580 |
},
|
| 581 |
{
|
| 582 |
+
"epoch": 12.48,
|
|
|
|
| 583 |
"learning_rate": 0.0003,
|
| 584 |
+
"loss": 8.0191,
|
| 585 |
"step": 6600
|
| 586 |
},
|
| 587 |
{
|
| 588 |
+
"epoch": 12.67,
|
|
|
|
| 589 |
"learning_rate": 0.0003,
|
| 590 |
+
"loss": 8.0091,
|
| 591 |
"step": 6700
|
| 592 |
},
|
| 593 |
{
|
| 594 |
+
"epoch": 12.85,
|
|
|
|
| 595 |
"learning_rate": 0.0003,
|
| 596 |
+
"loss": 7.9902,
|
| 597 |
"step": 6800
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 13.0,
|
| 601 |
+
"eval_accuracy": 0.2196923076923077,
|
| 602 |
+
"eval_loss": 6.872635841369629,
|
| 603 |
+
"eval_runtime": 4.943,
|
| 604 |
+
"eval_samples_per_second": 101.152,
|
| 605 |
+
"eval_steps_per_second": 12.745,
|
| 606 |
"step": 6877
|
| 607 |
},
|
| 608 |
{
|
|
|
|
| 612 |
"step": 6877
|
| 613 |
},
|
| 614 |
{
|
| 615 |
+
"epoch": 13.04,
|
|
|
|
| 616 |
"learning_rate": 0.0003,
|
| 617 |
+
"loss": 8.0141,
|
| 618 |
"step": 6900
|
| 619 |
},
|
| 620 |
{
|
| 621 |
+
"epoch": 13.23,
|
|
|
|
| 622 |
"learning_rate": 0.0003,
|
| 623 |
+
"loss": 8.0041,
|
| 624 |
"step": 7000
|
| 625 |
},
|
| 626 |
{
|
| 627 |
+
"epoch": 13.42,
|
|
|
|
| 628 |
"learning_rate": 0.0003,
|
| 629 |
+
"loss": 7.9641,
|
| 630 |
"step": 7100
|
| 631 |
},
|
| 632 |
{
|
| 633 |
+
"epoch": 13.61,
|
|
|
|
| 634 |
"learning_rate": 0.0003,
|
| 635 |
+
"loss": 7.9734,
|
| 636 |
"step": 7200
|
| 637 |
},
|
| 638 |
{
|
| 639 |
+
"epoch": 13.8,
|
|
|
|
| 640 |
"learning_rate": 0.0003,
|
| 641 |
+
"loss": 7.9893,
|
| 642 |
"step": 7300
|
| 643 |
},
|
| 644 |
{
|
| 645 |
+
"epoch": 13.99,
|
|
|
|
| 646 |
"learning_rate": 0.0003,
|
| 647 |
+
"loss": 7.8585,
|
| 648 |
"step": 7400
|
| 649 |
},
|
| 650 |
{
|
| 651 |
"epoch": 14.0,
|
| 652 |
+
"eval_accuracy": 0.2194871794871795,
|
| 653 |
+
"eval_loss": 6.823603630065918,
|
| 654 |
+
"eval_runtime": 4.416,
|
| 655 |
+
"eval_samples_per_second": 113.224,
|
| 656 |
+
"eval_steps_per_second": 14.266,
|
| 657 |
"step": 7406
|
| 658 |
},
|
| 659 |
{
|
|
|
|
| 663 |
"step": 7406
|
| 664 |
},
|
| 665 |
{
|
| 666 |
+
"epoch": 14.18,
|
|
|
|
| 667 |
"learning_rate": 0.0003,
|
| 668 |
+
"loss": 7.3595,
|
| 669 |
"step": 7500
|
| 670 |
},
|
| 671 |
{
|
| 672 |
+
"epoch": 14.37,
|
|
|
|
| 673 |
"learning_rate": 0.0003,
|
| 674 |
+
"loss": 7.3486,
|
| 675 |
"step": 7600
|
| 676 |
},
|
| 677 |
{
|
| 678 |
+
"epoch": 14.56,
|
|
|
|
| 679 |
"learning_rate": 0.0003,
|
| 680 |
+
"loss": 7.335,
|
| 681 |
"step": 7700
|
| 682 |
},
|
| 683 |
{
|
| 684 |
+
"epoch": 14.74,
|
|
|
|
| 685 |
"learning_rate": 0.0003,
|
| 686 |
+
"loss": 7.3724,
|
| 687 |
"step": 7800
|
| 688 |
},
|
| 689 |
{
|
| 690 |
+
"epoch": 14.93,
|
|
|
|
| 691 |
"learning_rate": 0.0003,
|
| 692 |
+
"loss": 7.3449,
|
| 693 |
"step": 7900
|
| 694 |
},
|
| 695 |
{
|
| 696 |
"epoch": 15.0,
|
| 697 |
+
"eval_accuracy": 0.1922051282051282,
|
| 698 |
+
"eval_loss": 7.199692249298096,
|
| 699 |
+
"eval_runtime": 4.4181,
|
| 700 |
+
"eval_samples_per_second": 113.17,
|
| 701 |
+
"eval_steps_per_second": 14.259,
|
| 702 |
"step": 7935
|
| 703 |
},
|
| 704 |
{
|
|
|
|
| 708 |
"step": 7935
|
| 709 |
},
|
| 710 |
{
|
| 711 |
+
"epoch": 15.12,
|
|
|
|
| 712 |
"learning_rate": 0.0003,
|
| 713 |
+
"loss": 7.3378,
|
| 714 |
"step": 8000
|
| 715 |
},
|
| 716 |
{
|
| 717 |
+
"epoch": 15.31,
|
|
|
|
| 718 |
"learning_rate": 0.0003,
|
| 719 |
+
"loss": 7.3525,
|
| 720 |
"step": 8100
|
| 721 |
},
|
| 722 |
{
|
| 723 |
+
"epoch": 15.5,
|
|
|
|
| 724 |
"learning_rate": 0.0003,
|
| 725 |
+
"loss": 7.3645,
|
| 726 |
"step": 8200
|
| 727 |
},
|
| 728 |
{
|
| 729 |
+
"epoch": 15.69,
|
|
|
|
| 730 |
"learning_rate": 0.0003,
|
| 731 |
+
"loss": 7.3455,
|
| 732 |
"step": 8300
|
| 733 |
},
|
| 734 |
{
|
| 735 |
+
"epoch": 15.88,
|
|
|
|
| 736 |
"learning_rate": 0.0003,
|
| 737 |
+
"loss": 7.3133,
|
| 738 |
"step": 8400
|
| 739 |
},
|
| 740 |
{
|
| 741 |
"epoch": 16.0,
|
| 742 |
+
"eval_accuracy": 0.18692307692307691,
|
| 743 |
+
"eval_loss": 6.745450496673584,
|
| 744 |
+
"eval_runtime": 4.218,
|
| 745 |
+
"eval_samples_per_second": 118.541,
|
| 746 |
+
"eval_steps_per_second": 14.936,
|
| 747 |
"step": 8464
|
| 748 |
},
|
| 749 |
{
|
|
|
|
| 753 |
"step": 8464
|
| 754 |
},
|
| 755 |
{
|
| 756 |
+
"epoch": 16.07,
|
|
|
|
| 757 |
"learning_rate": 0.0003,
|
| 758 |
+
"loss": 7.3153,
|
| 759 |
"step": 8500
|
| 760 |
},
|
| 761 |
{
|
| 762 |
+
"epoch": 16.26,
|
|
|
|
| 763 |
"learning_rate": 0.0003,
|
| 764 |
+
"loss": 7.3123,
|
| 765 |
"step": 8600
|
| 766 |
},
|
| 767 |
{
|
| 768 |
+
"epoch": 16.45,
|
|
|
|
| 769 |
"learning_rate": 0.0003,
|
| 770 |
+
"loss": 7.3081,
|
| 771 |
"step": 8700
|
| 772 |
},
|
| 773 |
{
|
| 774 |
+
"epoch": 16.64,
|
|
|
|
| 775 |
"learning_rate": 0.0003,
|
| 776 |
+
"loss": 7.3069,
|
| 777 |
"step": 8800
|
| 778 |
},
|
| 779 |
{
|
| 780 |
+
"epoch": 16.82,
|
|
|
|
| 781 |
"learning_rate": 0.0003,
|
| 782 |
+
"loss": 7.305,
|
| 783 |
"step": 8900
|
| 784 |
},
|
| 785 |
{
|
| 786 |
"epoch": 17.0,
|
| 787 |
+
"eval_accuracy": 0.18692307692307691,
|
| 788 |
+
"eval_loss": 6.745449066162109,
|
| 789 |
+
"eval_runtime": 4.2111,
|
| 790 |
+
"eval_samples_per_second": 118.735,
|
| 791 |
+
"eval_steps_per_second": 14.961,
|
| 792 |
"step": 8993
|
| 793 |
},
|
| 794 |
{
|
|
|
|
| 798 |
"step": 8993
|
| 799 |
},
|
| 800 |
{
|
| 801 |
+
"epoch": 17.01,
|
|
|
|
| 802 |
"learning_rate": 0.0003,
|
| 803 |
+
"loss": 7.3114,
|
| 804 |
"step": 9000
|
| 805 |
},
|
| 806 |
{
|
| 807 |
+
"epoch": 17.2,
|
|
|
|
| 808 |
"learning_rate": 0.0003,
|
| 809 |
+
"loss": 7.3226,
|
| 810 |
"step": 9100
|
| 811 |
},
|
| 812 |
{
|
| 813 |
+
"epoch": 17.39,
|
|
|
|
| 814 |
"learning_rate": 0.0003,
|
| 815 |
+
"loss": 7.3234,
|
| 816 |
"step": 9200
|
| 817 |
},
|
| 818 |
{
|
| 819 |
+
"epoch": 17.58,
|
|
|
|
| 820 |
"learning_rate": 0.0003,
|
| 821 |
+
"loss": 7.3929,
|
| 822 |
"step": 9300
|
| 823 |
},
|
| 824 |
{
|
| 825 |
+
"epoch": 17.77,
|
|
|
|
| 826 |
"learning_rate": 0.0003,
|
| 827 |
+
"loss": 7.5523,
|
| 828 |
"step": 9400
|
| 829 |
},
|
| 830 |
{
|
| 831 |
+
"epoch": 17.96,
|
|
|
|
| 832 |
"learning_rate": 0.0003,
|
| 833 |
+
"loss": 7.7463,
|
| 834 |
"step": 9500
|
| 835 |
},
|
| 836 |
{
|
| 837 |
"epoch": 18.0,
|
| 838 |
+
"eval_accuracy": 0.18697435897435896,
|
| 839 |
+
"eval_loss": 8.831918716430664,
|
| 840 |
+
"eval_runtime": 4.1023,
|
| 841 |
+
"eval_samples_per_second": 121.881,
|
| 842 |
+
"eval_steps_per_second": 15.357,
|
| 843 |
"step": 9522
|
| 844 |
},
|
| 845 |
{
|
|
|
|
| 849 |
"step": 9522
|
| 850 |
},
|
| 851 |
{
|
| 852 |
+
"epoch": 18.15,
|
|
|
|
| 853 |
"learning_rate": 0.0003,
|
| 854 |
+
"loss": 9.7591,
|
| 855 |
"step": 9600
|
| 856 |
},
|
| 857 |
{
|
| 858 |
+
"epoch": 18.34,
|
|
|
|
| 859 |
"learning_rate": 0.0003,
|
| 860 |
+
"loss": 10.4962,
|
| 861 |
"step": 9700
|
| 862 |
},
|
| 863 |
{
|
| 864 |
+
"epoch": 18.53,
|
|
|
|
| 865 |
"learning_rate": 0.0003,
|
| 866 |
+
"loss": 9.9571,
|
| 867 |
"step": 9800
|
| 868 |
},
|
| 869 |
{
|
| 870 |
+
"epoch": 18.71,
|
|
|
|
| 871 |
"learning_rate": 0.0003,
|
| 872 |
+
"loss": 9.9661,
|
| 873 |
"step": 9900
|
| 874 |
},
|
| 875 |
{
|
| 876 |
+
"epoch": 18.9,
|
|
|
|
| 877 |
"learning_rate": 0.0003,
|
| 878 |
+
"loss": 9.9696,
|
| 879 |
"step": 10000
|
| 880 |
},
|
| 881 |
{
|
| 882 |
"epoch": 19.0,
|
| 883 |
+
"eval_accuracy": 0.16923076923076924,
|
| 884 |
+
"eval_loss": 10.070154190063477,
|
| 885 |
+
"eval_runtime": 4.1088,
|
| 886 |
+
"eval_samples_per_second": 121.69,
|
| 887 |
+
"eval_steps_per_second": 15.333,
|
| 888 |
"step": 10051
|
| 889 |
},
|
| 890 |
{
|
|
|
|
| 894 |
"step": 10051
|
| 895 |
},
|
| 896 |
{
|
| 897 |
+
"epoch": 19.09,
|
|
|
|
| 898 |
"learning_rate": 0.0003,
|
| 899 |
+
"loss": 9.9832,
|
| 900 |
"step": 10100
|
| 901 |
},
|
| 902 |
{
|
| 903 |
+
"epoch": 19.28,
|
|
|
|
| 904 |
"learning_rate": 0.0003,
|
| 905 |
+
"loss": 9.9764,
|
| 906 |
"step": 10200
|
| 907 |
},
|
| 908 |
{
|
| 909 |
+
"epoch": 19.47,
|
|
|
|
| 910 |
"learning_rate": 0.0003,
|
| 911 |
+
"loss": 9.967,
|
| 912 |
"step": 10300
|
| 913 |
},
|
| 914 |
{
|
| 915 |
+
"epoch": 19.66,
|
|
|
|
| 916 |
"learning_rate": 0.0003,
|
| 917 |
+
"loss": 9.9739,
|
| 918 |
"step": 10400
|
| 919 |
},
|
| 920 |
{
|
| 921 |
+
"epoch": 19.85,
|
|
|
|
| 922 |
"learning_rate": 0.0003,
|
| 923 |
+
"loss": 9.9845,
|
| 924 |
"step": 10500
|
| 925 |
},
|
| 926 |
{
|
| 927 |
"epoch": 20.0,
|
| 928 |
+
"eval_accuracy": 0.16923076923076924,
|
| 929 |
+
"eval_loss": 10.070154190063477,
|
| 930 |
+
"eval_runtime": 4.4134,
|
| 931 |
+
"eval_samples_per_second": 113.292,
|
| 932 |
+
"eval_steps_per_second": 14.275,
|
| 933 |
"step": 10580
|
| 934 |
},
|
| 935 |
{
|
|
|
|
| 941 |
],
|
| 942 |
"logging_steps": 100,
|
| 943 |
"max_steps": 26450,
|
|
|
|
| 944 |
"num_train_epochs": 50,
|
| 945 |
"save_steps": 500,
|
| 946 |
+
"total_flos": 1.7870734670660895e+18,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 947 |
"trial_name": null,
|
| 948 |
"trial_params": null
|
| 949 |
}
|
checkpoint-10580/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b59ca157797b57867ac15a1da8ca60d782441a1506bb7cb43703c7748b856900
|
| 3 |
+
size 4728
|